Reproducible Benchmarking for Lung Nodule Detection and Malignancy Classification Across Multiple Low-Dose CT Datasets

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Tushar, Fakrul Islam, Wang, Avivah, Dahal, Lavsen, Samei, Ehsan, Harowicz, Michael R., Kalpathy-Cramer, Jayashree, Lafata, Kyle J., Tailor, Tina D., Rudin, Cynthia, Lo, Joseph Y.
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914315025514496
author Tushar, Fakrul Islam
Wang, Avivah
Dahal, Lavsen
Samei, Ehsan
Harowicz, Michael R.
Kalpathy-Cramer, Jayashree
Lafata, Kyle J.
Tailor, Tina D.
Rudin, Cynthia
Lo, Joseph Y.
author_facet Tushar, Fakrul Islam
Wang, Avivah
Dahal, Lavsen
Samei, Ehsan
Harowicz, Michael R.
Kalpathy-Cramer, Jayashree
Lafata, Kyle J.
Tailor, Tina D.
Rudin, Cynthia
Lo, Joseph Y.
contents Evaluation of artificial intelligence (AI) models for low-dose CT lung cancer screening is limited by heterogeneous datasets, annotation standards, and evaluation protocols, making performance difficult to compare and translate across clinical settings. We establish a public, reproducible multi-dataset benchmark for lung nodule detection and nodule-level cancer classification and quantify cross-dataset generalizability. Using the Duke Lung Cancer Screening (DLCS) dataset as a clinically curated development set, we evaluate performance across LUNA16/LIDC-IDRI, NLST-3D, and LUNA25. Detection models trained on DLCS and LUNA16 were evaluated externally on NLST-3D using free-response ROC analysis. For malignancy classification, we compared five strategies: randomly initialized ResNet50, Models Genesis, Med3D, a Foundation Model for Cancer Biomarkers, and a Strategic Warm-Start (ResNet50-SWS) approach pretrained using detection-derived candidate patches stratified by confidence. Performance was summarized using AUC with 95% confidence intervals and DeLong tests. Detection performance varied substantially by training dataset, with DLCS-trained models outperforming LUNA16-trained models on external NLST-3D evaluation (sensitivity at 2 false positives per scan: 0.72 vs. 0.64; p < 0.001). For malignancy classification, ResNet50-SWS achieved AUCs of 0.71 (DLCS), 0.90 (LUNA16), 0.81 (NLST-3D), and 0.80 (LUNA25), consistently matching or exceeding alternative pretraining strategies. These results demonstrate that dataset characteristics strongly influence lung cancer AI performance and highlight the need for transparent, multi-dataset benchmarking.
format Preprint
id arxiv_https___arxiv_org_abs_2405_04605
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Reproducible Benchmarking for Lung Nodule Detection and Malignancy Classification Across Multiple Low-Dose CT Datasets
Tushar, Fakrul Islam
Wang, Avivah
Dahal, Lavsen
Samei, Ehsan
Harowicz, Michael R.
Kalpathy-Cramer, Jayashree
Lafata, Kyle J.
Tailor, Tina D.
Rudin, Cynthia
Lo, Joseph Y.
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Image and Video Processing
Evaluation of artificial intelligence (AI) models for low-dose CT lung cancer screening is limited by heterogeneous datasets, annotation standards, and evaluation protocols, making performance difficult to compare and translate across clinical settings. We establish a public, reproducible multi-dataset benchmark for lung nodule detection and nodule-level cancer classification and quantify cross-dataset generalizability. Using the Duke Lung Cancer Screening (DLCS) dataset as a clinically curated development set, we evaluate performance across LUNA16/LIDC-IDRI, NLST-3D, and LUNA25. Detection models trained on DLCS and LUNA16 were evaluated externally on NLST-3D using free-response ROC analysis. For malignancy classification, we compared five strategies: randomly initialized ResNet50, Models Genesis, Med3D, a Foundation Model for Cancer Biomarkers, and a Strategic Warm-Start (ResNet50-SWS) approach pretrained using detection-derived candidate patches stratified by confidence. Performance was summarized using AUC with 95% confidence intervals and DeLong tests. Detection performance varied substantially by training dataset, with DLCS-trained models outperforming LUNA16-trained models on external NLST-3D evaluation (sensitivity at 2 false positives per scan: 0.72 vs. 0.64; p < 0.001). For malignancy classification, ResNet50-SWS achieved AUCs of 0.71 (DLCS), 0.90 (LUNA16), 0.81 (NLST-3D), and 0.80 (LUNA25), consistently matching or exceeding alternative pretraining strategies. These results demonstrate that dataset characteristics strongly influence lung cancer AI performance and highlight the need for transparent, multi-dataset benchmarking.
title Reproducible Benchmarking for Lung Nodule Detection and Malignancy Classification Across Multiple Low-Dose CT Datasets
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Image and Video Processing
url https://arxiv.org/abs/2405.04605